SNARC and SNAAC: Spatial-numeric association of response-codes and attentional cuing

Comprehending quantity and order information denoted by numbers are important abilities in everyday life in most modern societies. Better understanding of how numbers are psychologically represented can lead to better educational practices. Much psychological evidence suggests that numbers are represented in a spatial format, as in a “mental number line” in which smaller magnitudes are associated with left space and larger magnitudes with right space. This representation can lead to facilitation or interference for selection of a spatially oriented response, such as making a left- or right-hand key-press, or orienting attention to left or right sides of visual space. Two event-related potential (ERP) experiments were conducted to examine effects on response-selection and visuospatial attention resulting from associations between numerical magnitude and spatial coordinates of bimanual responses and attentional orienting. Experiment 1 examined the time course of response-selection processes as modulated by spatial-numeric associations of response codes (SNARC) in combination with expectancy, in a task of comparing the magnitude of two sequentially presented digits. In SNARC-compatible conditions, participants made a left-hand response to say the second digit was smaller, and a right-hand response to say it was larger, than the first digit. In SNARC-incompatible conditions participants made a right-hand response to say the second digit was smaller, and a left-hand response to say it was larger, than the first digit. The first digit to be compared gave advance information about the likely magnitude of the second digit to be compared, and thus about the likely response to be required. Behavioral response times (RTs) showed that interference from SNARC-incompatible response mapping was attenuated when the first digit gave complete information about the upcoming second digit and the required response. However, SNARC interference was still evident in ERP covert measures of mental processing. Lateralized readiness potentials (LRPs) showed evidence for SNARC interference at intermediate stages of processing, consistent with a limited amount of ERP research into SNARC effects, when response-cuing provided complete information in advance about the required response. Additionally the non-lateralized P300 showed clear SNARC effects in contrast to previous findings. Results are discussed in terms of a modified “dual-route” model, in which fast activation of task-irrelevant information competes for response-selection with slow activation of information related to task goals. Experiment 2 examined whether visuospatial attention shifts can be elicited automatically by numerical magnitude, by presenting task-irrelevant visual probes in left or right peripheral visual space following small- or large-magnitude digit pre-cues. Early visual ERPs showed a possible trend for enhancement after small-magnitude digits regardless of visual hemifield in which the probe appeared. Results across experiments suggest that spatialized number representation has different consequences within reference frames for either action or perception.